Question Ryzen 7 7700X runs very hot in games according to HWMonitor ?

[Barak9006]

I have new pc with new cooler Corsair Water Cooling H150 RGB 360mm
Case:Corsair iCUE 5000X RGB TG Smart Case Black
Mobo: Gigabyte X670E Aorus Pro X

At idle, the CPU runs fine at 50-60 Celsius, but when I open a game it becomes like this:

https://imgur.com/v7giYDm

Game used: Guild-Wars2

The air coming from my case is cold.
I got my computer checked twice to check if the cooler is seated correctly, and it is.


Thank you.

​

 

[Roland Of Gilead]

Hey there,

No, something is up. Idle at 50-60 is not okay. It should be in the low 30's. There must be something else at play. What thermal paste is used? Who checked your computer? Are all the fans on the rad working? You might expect games to be aorund 70-80c, but no more than that really.

Full load, you would expect high 80-90's.

 

[Barak9006]

Hey there,

No, something is up. Idle at 50-60 is not okay. It should be in the low 30's. There must be something else at play. What thermal paste is used? Who checked your computer? Are all the fans on the rad working? You might expect games to be aorund 70-80c, but no more than that really.

Full load, you would expect high 80-90's.

Right now what I did is, the mobo set cpu voltage at 1.4v. I set it on 1.2v and now still idle at 50, but now at full load it is 73-75. The one who checked my pc is the vendor technician. all fans are working. I dont know what thermal used I guess the one that comes with the cpu.

 

[Roland Of Gilead]

What bios are you running on your mobo? You can find this out using CPU-z.

 

[hotaru.hino]

I'm going to call this "expected behavior." And I'll leave my thoughts in bullet point form.

• One source of contention is that socket AM5 IHSes are thick.
  
  
  
  
  While copper is a good conductor of heat, it's an increased distance to travel. I've seen one post say this isn't a problem with math, but they only accounted for the thermal conductivity value (k-value), they didn't factor the thermal resistance value (r-value). Granted copper's r-value is still pretty low, but it might matter enough.
  
  Also there's evidence to suggest that grinding a millimeter or two can help with temperatures: https://www.tomshardware.com/news/g...-seemingly-lowers-temps-by-10-degrees-celsius
• At least one outlet reported scratching on their AIO cold plates from the IHS
• There's the point that the CCDs are smaller in size (70mm^2 compared to Zen 3's 83mm^2), but dissipate the same if not more power. This means more heat is concentrated on the die. This actually makes it harder to cool.
• Due to the position of the chiplets, the heat spreader will start off with uneven heating. See https://www.igorslab.de/en/amd-ryze...nalysis-temperatures-hotspots-and-problems/3/ .
  
  In response to this, Noctua made an offset mounting bracket to position the center of the cooler closer to the CCDs, to which they claim can reduce temperatures up to 3C.
• I also looked around the internet to see what other people were getting with regards to idle temps on the 7700X. Most of them were floating around 40-45C even with water cooling. So 50C on OP's end is a little warmer than what other people were reporting, but I would not expect the CPU to idle below 40C.
  
  Another thing to note is that Ryzen processors idle at a much higher power consumption than Intel. I haven't seen below 20W on a typical setup, whereas I have an Intel CPU that sits at ~6W on average for total package power when idling.
  
  
• Of note, Zen 4 CPUs are designed to hit their thermal limits first when running a load. So it's actually expected and by design for the CPU to operate close to or at its Tj_max temperature of 95C. From https://arstechnica.com/gadgets/202...ew-zen-4-starts-off-expensive-but-impressive/

For what it's worth, the AMD reviewers' guide says that this is safe, expected behavior:

With the new AM5 socket and higher TDP, most processors will run into a thermal wall before they hit a power wall. You will therefore see the Ryzen 7000 series, especially the higher core count variants, reside at TJMax (about 95 degrees Celsius for the Ryzen 7000 series) when running intense multithreaded workloads like Cinebench nt. This behavior is intended and by design.

It’s important to note TJMax is the max safe operating temperature—not the absolute max temperature. In the Ryzen 7000 Series, the processor is designed to run at TJMax 24/7 without risk of damage or deterioration. At 95 degrees it is not running hot, rather it will intentionally go to this temperature as much as possible under load because the power management system knows that this is the ideal way to squeeze the most performance out of the chip without damaging it.

 

[Barak9006]

What bios are you running on your mobo? You an find this out using CPU-z.

F30
5/22/24

 

[Barak9006]

I'm going to call this "expected behavior." And I'll leave my thoughts in bullet point form.

• One source of contention is that socket AM5 IHSes are thick.
  
  
  
  
  While copper is a good conductor of heat, it's an increased distance to travel. I've seen one post say this isn't a problem with math, but they only accounted for the thermal conductivity value (k-value), they didn't factor the thermal resistance value (r-value). Granted copper's r-value is still pretty low, but it might matter enough.
  
  Also there's evidence to suggest that grinding a millimeter or two can help with temperatures: https://www.tomshardware.com/news/g...-seemingly-lowers-temps-by-10-degrees-celsius
• At least one outlet reported scratching on their AIO cold plates from the IHS
• There's the point that the CCDs are smaller in size (70mm^2 compared to Zen 3's 83mm^2), but dissipate the same if not more power. This means more heat is concentrated on the die. This actually makes it harder to cool.
• Due to the position of the chiplets, the heat spreader will start off with uneven heating. See https://www.igorslab.de/en/amd-ryze...nalysis-temperatures-hotspots-and-problems/3/ .
  
  In response to this, Noctua made an offset mounting bracket to position the center of the cooler closer to the CCDs, to which they claim can reduce temperatures up to 3C.
• I also looked around the internet to see what other people were getting with regards to idle temps on the 7700X. Most of them were floating around 40-45C even with water cooling. So 50C on OP's end is a little warmer than what other people were reporting, but I would not expect the CPU to idle below 40C.
  
  Another thing to note is that Ryzen processors idle at a much higher power consumption than Intel. I haven't seen below 20W on a typical setup, whereas I have an Intel CPU that sits at ~6W on average for total package power when idling.
  
  
• Of note, Zen 4 CPUs are designed to hit their thermal limits first when running a load. So it's actually expected and by design for the CPU to operate close to or at its Tj_max temperature of 95C. From https://arstechnica.com/gadgets/202...ew-zen-4-starts-off-expensive-but-impressive/

I should be getting low temp like other reported since to my knowledge, I bought decent cooler.

I heard that 7000 series having a high idle temp. But I have friends with the 7800x3d which are having lower than 40 Celsius out of the box.

 

[hotaru.hino]

I should be getting low temp like other reported since to my knowledge, I bought decent cooler.

I heard that 7000 series having a high idle temp. But I have friends with the 7800x3d which are having lower than 40 Celsius out of the box.

Who are the "others" that reported this and did they have the same setup as you? Cooling depends on a lot of factors.

I mean, you can try and re-paste the cooler and remount it, but if that's not working, then there could be a multitude of other things going on.

 

[Barak9006]

Who are the "others" that reported this and did they have the same setup as you? Cooling depends on a lot of factors.

I mean, you can try and re-paste the cooler and remount it, but if that's not working, then there could be a multitude of other things going on.

its same setup though different cpu(still on the 7000 series) some even worse cooler than mine.

 

[hotaru.hino]

its same setup though different cpu(still on the 7000 series) some even worse cooler than mine.

Same setup as in:

• Same cooler
• Same thermal paste
• Same case
• Same fans and fan curves
• Similar place where the computer is kept
• Similar environmental temperature
• Similar idle workload

And really this could still be as simple as something non-ideal happened during the cooler installation. If the person used the old method of "rice grain dollop of thermal paste in the middle", that probably won't spread to provide enough coverage across the IHS, assuming they didn't use the stuff that came on the AIO.

 

[Barak9006]

Same setup as in:

• Same cooler
• Same thermal paste
• Same case
• Same fans and fan curves
• Similar place where the computer is kept
• Similar environmental temperature
• Similar idle workload

And really this could still be as simple as something non-ideal happened during the cooler installation. If the person used the old method of "rice grain dollop of thermal paste in the middle", that probably won't spread to provide enough coverage across the IHS, assuming they didn't use the stuff that came on the AIO.

I get your point. Some stuff are beyond my control, but bottom line is 50-55 degrees is fine but can still able to be reduced by multitude of things you mentioned.

 

[kerberos_20]

cpu is throttling, 65-75watt power draw is low, cpu is able to go 105+watts..so his cooler isnt working, maybe pump dead

 

[Roland Of Gilead]

I also looked around the internet to see what other people were getting with regards to idle temps on the 7700X. Most of them were floating around 40-45C even with water cooling. So 50C on OP's end is a little warmer than what other people were reporting, but I would not expect the CPU to idle below 40C.

Hmmm. I don't know. The reviews I've seen have idle in mid 30's, which is what I'd expect. However, I would agree that there are many reasons why the temps could be higher, including the cooler, paste etc as you've mentioned. From a software point of view there could be many processes running taking up CPU cycles.

F30
5/22/24

F30 is the most recent one, I can see so that's good. When you updated the bios, did you clear CMOS (CLRTC) afterwards?

 

[Barak9006]

cpu is throttling, 65-75watt power draw is low, cpu is able to go 105+watts..so his cooler isnt working, maybe pump dead

hey,

I revert to the voltage 1.3v from 1.2v to show you current stats:

This is on idle(can see load there)

I dont know if fans RPM is indication that it works or not.

 

[kerberos_20]

hey,

I revert to the voltage 1.3v from 1.2v to show you current stats:

This is on idle(can see load there)

I dont know if fans RPM is indication that it works or not.

that doesnt tell much, as there should be fan labeled "pump" from icue

if pump missing, than recheck pump connector if its sitting properly in that corsair adapter

 

[Hotrod2go]

Is this cpu running PBO enabled? & if it is have you tuned the curve optimizer? This chip looks like one that demands a high VID, controlling the VID is the key to lowering temps because if you have taken this PC to a technician (presuming they are competent) & they checked the hardware installation then this is definately one area you need to pay attention too. Negative values in the curve optimizer profile are what your after to reduce VID demands. But you'll have to experiment & test with suitable CPU stress apps to see what works & what doesn't.
If manually adjusting PBO limits is too technical, your mobo should have built in PBO profiles in the bios. So check them out.

 

[kerberos_20]

If manually adjusting PBO limits is too technical, your mobo should have built in PBO profiles in the bios. So check them out.

but how does adjusting PBO limits would help? his picture shows PPT 75watts (54%), EDC 80A (40%), TDC 44A (40%), meaning he can turn off PBO overclock and it wont do a thing since hes at ~50% PBO
yet CPU is melting with 360mm AIO

another picture sshows 7% cpu load at 70C...thats not something you can overcome with PBO...

cant be isue with VID either, since amperage and wattage is low

 

[Barak9006]

that doesnt tell much, as there should be fan labeled "pump" from icue

if pump missing, than recheck pump connector if its sitting properly in that corsair adapter

Yeah im gonna send it back to the retailer and ask them to replace the cooler. Just a question, is that a good cooler or not? is there a problem with it? cause tried google and it seems a good cooler...

 

[Hotrod2go]

but how does adjusting PBO limits would help? his picture shows PPT 75watts (54%), EDC 80A (40%), TDC 44A (40%), meaning he can turn off PBO overclock and it wont do a thing since hes at ~50% PBO
yet CPU is melting with 360mm AIO

another picture sshows 7% cpu load at 70C...thats not something you can overcome with PBO...

cant be isue with VID either, since amperage and wattage is low

PBO is by default on auto unless the user changes it. On auto it will boost in games, that is how AMD have designed it. Adjusting VID obviously can increase or decrease heat, even at idle. Also, I place no faith in 3rd part tools to indicate technical details like thermals, etc unless its through HWiNFO. (There is a reason why many major vendors use HWiNFO above other 3rd party info tools - its more accurate than other tools out there on the market & that goes for even mobo vendors software.) That screenshot above from HWiNFO does not show VID values.
Controlling the VID is a necessary tool for effective thermal management, regardless of ambient temps or cooling solution. The OP may have a chip that is thirsty for high VID, unless extensive testing is carried out, that can never be ruled out. This phenomena is also known as "silicon lottery".